The broad long-term goal of this research program is to develop and validate a World Wide Web-based program to educate cancer registrars to code comorbidity information for inclusion into cancer registries. This research is important because if cancer registrars learn to code comorbidity accurately, consistently, and efficiently using a Web-based education program, then comorbidity information can be easily included as a standard oncology data element. Because of the demonstrated importance of comorbidity in cancer statistics, inclusion of this information will improve the accuracy of survival statistics and the evaluation of treatment effectiveness. These improvements will lead to a decrease in mortality and better quality of life for cancer patients. The specific aims of this continuation project are: 1. To develop a web-based education program to train cancer registrars to code comorbidity based on the teaching program developed and validated during previous research (R25CA68304-01, 02) 2. To assess the educational utility of this program This project will be a two-year, prospective, longitudinal study with two distinct phases. Phase I will include the development of a web-based education program to train cancer registrars to code comorbidity. The current paper-based Comorbidity Education Program will serve as the framework for the Web-Based Comorbidity Program. The program will follow the general format of the previous three-day training session and will be broken down into three parts: Learning the Coding Protocol, Knowledge Application to Chart Review, and Performance Assessment. Macromedia Course builder for Dreamweaver 3 will be used to develop the program. After the development of the program, cancer registars at local area hospital will view the program for its technical and educational merit. Based on the findings, the developers will make modifications. Phase 2 will assess the educational utility of this program. Five beta-test hospitals around the United States will participate. Registrars at these centers will learn to code comorbidity using the Web-based Program. The null hypothesis is that there will be no difference in the ability of registars to code comorbidity from the medical records of patients with cancer based on whether they received education from a trained research assistant or from the web-based computer program. The reliability of the registrars will be compared with the weighted kappa statistic and the validity will be compared with sensitivity and specificity. The traditional method for professional and continuing education for cancer registars is to attend courses sponsored by national or state professional organizations at their regular meetings. An alternative, low-cost option to train registrars to code comorbidity and provide updated information is to use the Internet via the World Wide Web.